检索结果-内蒙古大学图书馆

Unsupervised machine learning approach for tailoring educational content to individual student weaknesses

High-Confidence Computing 2024年第4期4卷 91-104页

作者： Shabab Intishar Rahman Shadman Ahmed Tasnim Akter Fariha Ammar Mohammad Muhammad Nayeem Mubasshirul Haque Sriram Chellappan Jannatun Noor Department of Computer Science and Engineering BRAC UniversityDhaka 1212Bangladesh Department of Computer Science and Engineering University of South FloridaFL 33620USA Computing for Sustainability and Social Good(C2SG)Research Group Department of Computer Science and EngineeringSchool of Data and SciencesBRAC UniversityDhaka 1212Bangladesh

By analyzing data gathered through Online Learning(OL)systems,data mining can be used to unearth hidden relationships between topics and trends in student ***,in this paper,we show how data mining techniques such as clustering and association rule algorithms can be used on historical data to develop a unique recommendation system *** our implementation,we utilize historical data to generate association rules specifically for student test marks below a threshold of 60%.By focusing on marks below this threshold,we aim to identify and establish associations based on the patterns of weakness observed in the past ***,we leverage K-means clustering to provide instructors with visual representations of the generated *** strategy aids instructors in better comprehending the information and associations produced by the *** clustering helps visualize and organize the data in a way that makes it easier for instructors to analyze and gain insights,enabling them to support the verification of the relationship between *** can be a useful tool to deliver better feedback to students as well as provide better insights to instructors when developing their *** paper further shows a prototype implementation of the above-mentioned concepts to gain opinions and insights about the usability and viability of the proposed system.

关键词： Recommendation system Clustering F-P growth Apriori Associative pattern E-learning sphere Online learning

来源：评论

学校读者我要写书评

暂无评论

AVF-Lip: High-Fidelity Talking Face Generation via Audio-Visual Fusion 7

AVF-Lip: High-Fidelity Talking Face Generation via Audio-Vis...

引用

7th International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2024

作者： Wu, Mengduo Zhao, Sirui Wu, Tong Xu, Yifan Xu, Tong Chen, Enhong University of Science and Technology of China School of Computer Science and Technology Hefei China University of Science and Technology of China School of Data Science Hefei China

ISBN: (纸本)9798350350890

Talking face generation aims to create realistic facial videos with lips precisely synchronized to the input audio, finding broad applications in human-computer interaction, virtual avatars, and video conferencing. Despite significant advancements in previous works, existing methods continue to face challenges in achieving accurate lip synchronization and high visual quality. To address these challenges, we propose a talking face generation framework named AVF-Lip that supports higher resolution and accurate lip synchronization. Specifically, an audio-visual fusion module (AVFM) is first customized to improve lip synchronization performance, and this is further enhanced by introducing a dynamic and fine-grained face mask. Subsequently, to address the problem of lip distortion in high-resolution facial images, we design a set of loss functions based on face sketches to enhance the generation of the lip areas. Extensive experiments demonstrate that our AVF-Lip outperforms state-of-the-art methods in lip synchronization and delivers competitive visual quality, positioning it as a promising solution for high-resolution and high-fidelity talking face generation. Project page: https://***/ © 2024 IEEE.

关键词： Video conferencing

来源：评论

学校读者我要写书评

暂无评论

CMDM: A Control Motion Diffusion Model for 2D Digital Human Motion Video Generation 7

CMDM: A Control Motion Diffusion Model for 2D Digital Human ...

引用

7th International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2024

作者： Wu, Tong Zhao, Sirui Jin, Siyuan Xu, Tong Chen, Enhong School of Data Science University of Science and Technology of China Hefei China School of Computer Science and Technology University of Science and Technology of China Hefei China

ISBN: (纸本)9798350350890

In recent years, 2D digital human motion generation (DHMG) is becoming increasingly crucial for many areas such as virtual live broadcasting and film production. Although a lot of effort has been invested in DHMG, the complexity and diversity of human movements make it difficult to generate motions for arbitrary poses, and it is challenging to avoid distortion in the generated hand regions. To address these challenges, in this paper, we propose Control Motion Diffusion Model (CMDM), a diffusion based method for generating 2D digital human motion videos with arbitrary given poses. In CMDM, a novel Pose Control Module (PCM) is customized to fully encode pose and character information, generating multi-level and multi-resolution embedding, thereby ensuring that the model can learn more precise human motion features. In addition, data augmentation and VAE fine-tuning training strategies are adopted to decouple character and pose features, improving the model's ability to generate unseen poses. Finally, comprehensive experiments have been completed to demonstrate the effectiveness of the proposed method, and the proposed CMDM has achieved significant improvement in both subjective and objective evaluations. © 2024 IEEE.

关键词： Digital elevation model

来源：评论

学校读者我要写书评

暂无评论

An Adaptive DoH Encrypted Tunnel Detection Method Based on Contrastive Learning

引用

IEEE Internet of Things Journal 2025年第13期12卷 25936-25950页

作者： Tong, Jiacheng Zhao, Yilin Jin, Chongju Chen, Wei Zhang, Yiting Wu, Lifa Nanjing University of Posts and Telecommunications School of Computer Science Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing210003 China

The percentage of encrypted network traffic has constantly increased as network security has been continuously improved. Attackers can, however, utilize encrypted DNS over HTTPS (DoH) to conceal their malicious traffic, which makes it more difficult to identify malicious tunnels. To address this issue, we first examine the encryption features of DoH tunnel traffic. Due to the incapability of current detection techniques to properly fuse traffic attributes, a fusion learning-based method is proposed to detect DoH encrypted tunnel traffic. At the same time, we discover that the DoH traffic samples may exhibit concept drift. As a result, we present a concept drift detection approach based on a contrastive sparse autoencoder. In addition to the above method, a model retraining strategy is also suggested to improve the model’s capacity to identify new DoH encrypted tunnel traffic while reducing its reliance on expert label data. This strategy involves incrementally training the model using as few samples as possible. Experiments demonstrate that the proposed method can significantly enhance detection performance. When 7% of drift samples are used during incremental training, the detection accuracy of the model recovers from 74.02% to 99.96%. © 2014 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Scaling Up Multi-Agent Reinforcement Learning via Graph Decomposition Invariant Network 5

Scaling Up Multi-Agent Reinforcement Learning via Graph Deco...

引用

5th International Seminar on Artificial Intelligence, Networking and Information Technology, AINIT 2024

作者： Fu, Hongyi Ji, Jianmin School of Data Science University of Science and Technology of China Hefei China School of Computer Science and Technology University of Science and Technology of China Hefei China

ISBN: (纸本)9798350385557

This paper focuses on the dilemma faced by largescale multi-agent systems. With the increase of agent size, the policy search space grows exponentially, reaching the task complexity that is difficult for the current mainstream multiagent algorithms to deal with, so that the optimal strategy is converged in large-scale multi-agent systems. In order to make the learning process of large-scale multi-agent tasks more smooth, many researchers have introduced the method of course learning, setting the scale from small to large courses, compared with learning training from scratch, this way can continue to inherit the knowledge learned in the smaller scale tasks, making the training smoother. However, in order to realize the multiagent course learning from small to large scale, we face the following difficulties. First, the neural network structure we designed can adapt to the observation input of agents of different scales, rather than only applicable to a fixed scale. Second, in the multi-stage course learning, due to the network's inefficient representation of multi-agent systems, the knowledge learned in small-scale courses cannot be effectively utilized in large-scale courses, and even the problem of catastrophic forgetting occurs in the neural network. Specifically, we view multi-agent systems from a graph perspective and derive the Graph Decomposition Invariant (GDI) for the first time. We prove that neural network architectures satisfying the GDI are of great help in realizing efficient course learning. We design a model architecture called GDI model suitable for multi-agent curriculum learning, which satisfies the graph decomposition invariant proposed by us. Finally, we designed small-scale to large-scale multi-agent tasks on two experimental platforms, StarCraft and Neural MMO, and conducted a large number of comparative experiments, which verified that our proposed GDI model has a great promotion effect on course learning. © 2024 IEEE.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Spatio-Temporal Location Recommendation for Urban Facility Placement via Graph Convolutional and Recurrent Networks

引用

Journal of computer science & Technology 2024年第6期39卷 1419-1440页

作者： Pu Wang Jian-Jiang Lu Wei Chen Peng-Peng Zhao Lei Zhao Institute of Artificial Intelligence School of Computer Science and TechnologySoochow UniversitySuzhou 215006China Department of Data Resource and Information Development Soochow UniversitySuzhou 215006China

The ability to recommend candidate locations for service facility placement is crucial for the success of urban planning. Whether a location is suitable for establishing new facilities is largely determined by its potential popularity. However, it is a non-trivial task to predict popularity of candidate locations due to three significant challenges: 1) the spatio-temporal behavior correlations of urban dwellers, 2) the spatial correlations between candidate locations and existing facilities, and 3) the temporal auto-correlations of locations themselves. To this end, we propose a novel semi-supervised learning model, Spatio-Temporal Graph Convolutional and Recurrent Networks (STGCRN), aiming for popularity prediction and location recommendation. Specifically, we first partition the urban space into spatial neighborhood regions centered by locations, extract the corresponding features, and develop the location correlation graph. Next, a contextual graph convolution module based on the attention mechanism is introduced to incorporate local and global spatial correlations among locations. A recurrent neural network is proposed to capture temporal dependencies between locations. Furthermore, we adopt a location popularity approximation block to estimate the missing popularity from both the spatial and temporal domains. Finally, the overall implicit characteristics are concatenated and then fed into the recurrent neural network to obtain the ultimate popularity. The extensive experiments on two real-world datasets demonstrate the superiority of the proposed model compared with state-of-the-art baselines.

关键词： location recommendation popularity prediction spatio-temporal correlation deep learning

来源：评论

学校读者我要写书评

暂无评论

AIGLLM: An Action Instruction Generation Method with Visually Enhanced LLM 7

AIGLLM: An Action Instruction Generation Method with Visuall...

引用

7th International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2024

作者： Xia, Yudong Zhao, Sirui Wu, Tong Tang, Huaying Xu, Tong University of Science and Technology of China School of Data Science Hefei China University of Science and Technology of China School of Computer Science and Technology Hefei China

ISBN: (纸本)9798350350890

Action Instruction Generation (AIG) is a key step in semantic-driven action sequence generation, which aims to embed text instructions describing specific actions into the driving text. AIG is a challenging task that requires accurately understanding the potential action semantics contained in the driving text. Besides, the challenge is exacerbated by the lack of paired text-action instructions data. To address these challenges and motivated by the powerful capabilities of LLMs in text generation, we propose AIGLLM, a novel AIG framework with visually enhanced Large Language Model (LLM). Specifically, we perform low-rank adaptation (LoRA) on the off-the-shelf LLM to enable the model understand action semantics and generate action instructions at appropriate positions within the original text sequence. Particularly, to address the current lack of data that can be used to fine-tune LLMs for AIG, we construct a dataset containing 12,000 text-action instruction pairs and 15,000 image-action instruction pairs. Subsequently, a visual module is designed to enable the LLM to learn action knowledge from external visual signals of actions, thereby enhancing the model's understanding of action semantics. Finally, we conduct extensive experiments, where the experimental results and evaluations demonstrate that our method has an impressive key benefit: improved accurate action instruction embedding. © 2024 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Efficacy of Latent Variables in Mitigating data Poisoning Attacks in the Context of Bayesian Networks:An Empirical Study

引用

computer Modeling in Engineering & sciences 2024年第5期139卷 1635-1654页

作者： Shahad Alzahrani Hatim Alsuwat Emad Alsuwat Independent Researcher Specializing in Data Security and Privacy Taif26571Saudi Arabia Department of Computer Science College of Computer and Information SystemsUmm Al-Qura UniversityMakkah24382Saudi Arabia Department of Computer Science College of Computers and Information TechnologyTaif UniversityTaif26571Saudi Arabia

Bayesian networks are a powerful class of graphical decision models used to represent causal relationships among ***,the reliability and integrity of learned Bayesian network models are highly dependent on the quality of incoming data *** of the primary challenges with Bayesian networks is their vulnerability to adversarial data poisoning attacks,wherein malicious data is injected into the training dataset to negatively influence the Bayesian network models and impair their *** this research paper,we propose an efficient framework for detecting data poisoning attacks against Bayesian network structure learning *** framework utilizes latent variables to quantify the amount of belief between every two nodes in each causal model over *** use our innovative methodology to tackle an important issue with data poisoning assaults in the context of Bayesian *** regard to four different forms of data poisoning attacks,we specifically aim to strengthen the security and dependability of Bayesian network structure learning techniques,such as the PC *** doing this,we explore the complexity of this area and offer workablemethods for identifying and reducing these sneaky ***,our research investigates one particular use case,the“Visit to Asia Network.”The practical consequences of using uncertainty as a way to spot cases of data poisoning are explored in this inquiry,which is of utmost *** results demonstrate the promising efficacy of latent variables in detecting and mitigating the threat of data poisoning ***,our proposed latent-based framework proves to be sensitive in detecting malicious data poisoning attacks in the context of stream data.

关键词： Bayesian networks data poisoning attacks latent variables structure learning algorithms adversarial attacks

来源：评论

学校读者我要写书评

暂无评论

Towards Long-Tailed Recognition for Graph Classification via Collaborative Experts

引用

IEEE Transactions on Big data 2023年第6期9卷 1683-1696页

作者： Yi, Si-Yu Mao, Zhengyang Ju, Wei Zhou, Yong-Dao Liu, Luchen Luo, Xiao Zhang, Ming School of Statistics and Data Science Nankai University Tianjin300071 China School of Computer Science Peking University Beijing100871 China Department of Computer Science University of California Los AngelesCA90095 United States

Graph classification, aiming at learning the graph-level representations for effective class assignments, has received outstanding achievements, which heavily relies on high-quality datasets that have balanced class distribution. In fact, most real-world graph data naturally presents a long-tailed form, where the head classes occupy much more samples than the tail classes, it thus is essential to study the graph-level classification over long-tailed data while still remaining largely unexplored. However, most existing long-tailed learning methods in visions fail to jointly optimize the representation learning and classifier training, as well as neglect the mining of the hard-to-classify classes. Directly applying existing methods to graphs may lead to sub-optimal performance, since the model trained on graphs would be more sensitive to the long-tailed distribution due to the complex topological characteristics. Hence, in this paper, we propose a novel long-tailed graph-level classification framework via Collaborative Multi-expert Learning (CoMe) to tackle the problem. To equilibrate the contributions of head and tail classes, we first develop balanced contrastive learning from the view of representation learning, and then design an individual-expert classifier training based on hard class mining. In addition, we execute gated fusion and disentangled knowledge distillation among the multiple experts to promote the collaboration in a multi-expert framework. Comprehensive experiments are performed on seven widely-used benchmark datasets to demonstrate the superiority of our method CoMe over state-of-the-art baselines. © 2023 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Facial Expression Recognition with DToF Sensing

Facial Expression Recognition with DToF Sensing

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Li, Chengxiao Zhang, Xie Wu, Chenshu Department of Computer Science School of Computing and Data Science The University of Hong Kong Hong Kong

ISBN: (纸本)9798350368741

Facial Expression Recognition (FER) is crucial for understanding human emotions, with applications spanning from mental health assessment to marketing recommendation systems. However, existing camera-based methods raise privacy concerns, while RF-based approaches suffer from limited environmental generalizability and high cost. In this work, we propose ToFace, a FER system leveraging a low-cost (4.8$) Direct Time-of-Flight (DToF) sensor that has been available on commodity smartphones. This sensor provides an extremely low-resolution 8 × 8 depth map and a clear Field of View (FoV), significantly mitigating privacy concerns while avoiding the impact of ambient objects. Despite the benefits, the low-resolution depth map introduces significant challenges for precise expression recognition due to limited facial structure information. We first develop a physical model to extract additional spatial information from the intermediate sensor output, i.e., the transient histograms. We then propose a physics-integrated neural network to reconstruct a facial structure map comprising both depth and orientation for accurate expression recognition. We conduct real-world experiments with 12 users and compare our model with several baselines. The results demonstrate that ToFace achieves the highest recognition accuracy of 75%. © 2025 IEEE.

关键词： direct time-of-flight sensor facial expression recognition internet of things sensing AI

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：